You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
@@ -87,7 +87,7 @@ Zero codemap timeouts. Plain-arm agents hit the 300-second hard timeout on sever
87
87
>
88
88
> codemap injects a rich dependency graph into every agent prompt. On weaker models or tasks with large blast-radius graphs, this extra context can overwhelm the model and cause it to fall back to grep-heavy loops — performing *worse* than plain arm. The benchmark labels this failure mode `degenerate_grep_loop`.
89
89
>
90
-
> Good integration requires three things: (1) **skill-first protocol** — the agent calls `/codemap:query` before any Grep/Glob; (2) **bounded call budget** — max 3 codemap queries per task; (3) **hard stop on `exhaustive: true`** — when the index says the list is complete, write the answer immediately, no more tool calls. Skipping any of these — especially ignoring the exhaustive flag — is the primary cause of regressions that flip the codemap benefit into a liability. Use `/codemap:integration init` to wire integration correctly rather than injecting context manually.
90
+
> Good integration requires three things: (1) **skill-first protocol** — the agent calls `/codemap:query-code` before any Grep/Glob; (2) **bounded call budget** — max 3 codemap queries per task; (3) **hard stop on `exhaustive: true`** — when the index says the list is complete, write the answer immediately, no more tool calls. Skipping any of these — especially ignoring the exhaustive flag — is the primary cause of regressions that flip the codemap benefit into a liability. Use `/codemap:integration init` to wire integration correctly rather than injecting context manually.
@@ -230,7 +230,7 @@ Each check prints `✓`, `✗`, or `⚠` with a one-line remediation hint if nee
230
230
231
231
Interactive onboarding for the current project:
232
232
233
-
1. Builds the index if it is missing (offers to run `/codemap:scan`)
233
+
1. Builds the index if it is missing (offers to run `/codemap:scan-codebase`)
234
234
2. Discovers all installed skills and agents across all plugins
235
235
3. Scores candidates by value tier (High / Medium / Low / Skip) based on whether structural context would help them
236
236
4. Presents a recommendation table and asks which to wire in
@@ -275,9 +275,9 @@ run `scan-query central --top 5` (and `scan-query rdeps <target_module>` when a
275
275
**before** any Glob/Grep exploration for structural information. Skip silently if the index is absent.
276
276
```
277
277
278
-
### scan
278
+
### scan-codebase
279
279
280
-
**Trigger**: `/codemap:scan`
280
+
**Trigger**: `/codemap:scan-codebase`
281
281
282
282
Builds the structural index by running `ast.parse` across every `.py` file in the project. Writes the index to `.cache/scan/<project>.json`. Reports how many modules were indexed, how many were degraded (parse errors), and which five modules have the highest blast radius.
283
283
@@ -303,28 +303,28 @@ Run a full scan once when you first set up the project. After that, `--increment
Copy file name to clipboardExpand all lines: plugins/codemap/skills/integration/SKILL.md
+8-8Lines changed: 8 additions & 8 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -14,7 +14,7 @@ Two modes: use `init` first-time to onboard, then `check` regularly to verify. D
14
14
-**`check`** — fast diagnostic: finds `scan-query`, verifies index exists and fresh, runs smoke test, audits which skill files have injection block. Prints `✓`/`✗`/`⚠` per check with one-line remediation hints. Pure bash — no model reasoning needed for happy path.
15
15
-**`init`** — interactive onboarding: builds index if missing, discovers all installed skills and agents, scores by how much codemap would help, presents recommendation table, asks which to wire in, inserts correct injection block into each selected file.
16
16
17
-
NOT for: building or rebuilding index (use `/codemap:scan`); running structural query (use `/codemap:query`).
17
+
NOT for: building or rebuilding index (use `/codemap:scan-codebase`); running structural query (use `/codemap:query-code`).
18
18
19
19
Arguments: `check` (no args) or `init [--approve]` — `--approve` auto-applies all ★ recommendations non-interactively.
20
20
@@ -71,7 +71,7 @@ if [ "$C1_STATUS" = "failed" ]; then
71
71
fi
72
72
# PROJ/INDEX resolution — also used in Step I1 (init mode); keep in sync
printf" → Re-run /codemap:scan to rebuild index\n"
127
+
printf" → Re-run /codemap:scan-codebase to rebuild index\n"
128
128
else
129
129
printf"✓ smoke test: index valid (mtime-age=%sh)\n""$AGE"
130
130
if [ "$STALE"="true" ];then
131
-
printf" ⚠ Index older than freshness threshold — run /codemap:scan to update\n"
131
+
printf" ⚠ Index older than freshness threshold — run /codemap:scan-codebase to update\n"
132
132
fi
133
133
fi
134
134
```
@@ -170,14 +170,14 @@ Use `AskUserQuestion`:
170
170
No codemap index found for project: $PROJ
171
171
172
172
a) Build now ★ — scans all .py files via ast.parse (Python only), <60s on most projects
173
-
b) Skip — I'll run /codemap:scan later (recommendations will be generic, no module-count weighting)
173
+
b) Skip — I'll run /codemap:scan-codebase later (recommendations will be generic, no module-count weighting)
174
174
```
175
175
176
176
If **a** (or auto-approved): verify binary exists first, then run scanner:
177
177
178
178
```bash
179
179
# timeout: 5000
180
-
[ -x"${CLAUDE_PLUGIN_ROOT:-plugins/codemap}/bin/scan-index" ] || { printf"✗ scan-index not found at ${CLAUDE_PLUGIN_ROOT:-plugins/codemap}/bin/scan-index\nTry: /codemap:scan to install and rebuild.\n";exit 1; }
180
+
[ -x"${CLAUDE_PLUGIN_ROOT:-plugins/codemap}/bin/scan-index" ] || { printf"✗ scan-index not found at ${CLAUDE_PLUGIN_ROOT:-plugins/codemap}/bin/scan-index\nTry: /codemap:scan-codebase to install and rebuild.\n";exit 1; }
0 commit comments